Perceived versus objective sleep quality in long-stay hospitalised older adults-a mixed methods study.

BACKGROUND: Hospital inpatients experience substantial sleep problems that have been linked with worse health outcomes, poor quality of life and the post-hospital syndrome. However, little is known about assessing sleep issues in older hospitalised patients. OBJECTIVE: To conduct an in-depth investigation on hospitalised older adults' sleep challenges and methods of sleep assessment. DESIGN: Cross-sectional observational study. SETTING: Public hospital inpatient unit. SUBJECTS: Long-stay hospitalised older adults. METHODS: Data were collected using validated sleep questionnaires, actigraphy devices and qualitative interviews. Quantitative data were analysed with descriptive statistics, multiple logistic regression and Cohen's Kappa. Qualitative data were analysed with qualitative content analysis; findings compared to the quantitative assessments. RESULTS: We collected data on 33 older long-stay hospital inpatients, who were mean (SD) 80.2(7.4) years old, 57.6% female and were hospitalised following stroke, medical illness and orthopaedic fracture. Mean (SD) total sleep time and actigraphic sleep efficiency were 480.6(73.6) minutes and 81.5(11.2)%, respectively. About, 57.6% were poor sleepers (Pittsburgh Sleep Quality Index [PSQI]) and 30.8% had indicators of clinical depression/low quality of life (WHO-5 well-being index). Three main themes were identified: "sleep assessment"; "factors that affect sleep"; "expectations of sleep". Bad sleepers were more likely to feel a lack of control over their sleep, while good sleepers spoke about the ability to adjust and accept their circumstances. CONCLUSIONS: We found high levels of sleep problems and identified substantial discrepancies between the validated sleep questionnaire and qualitative response data. Our findings indicate that standard assessment tools, such as PSQI, may not be suitable to assess sleep in hospitalised older adults and call for further investigations to build more appropriate methods. Further exploring psychological factors and expectations could potentially lead to novel interventions to improve sleep in this setting.